Stern scare blunted by the figures

The report on climate change by Nicholas Stern and the British Government has sparked publicity and scary headlines across the world. Much attention has been devoted to Stern's core argument that the price of inaction would be extraordinary and the cost of action modest.

Unfortunately, this claim falls apart when one reads the 700-page tome. Despite using many good references, the Stern Review on the Economics of Climate Change is selective and its conclusion flawed. Its fear-mongering arguments have been sensationalised, which is ultimately only likely to make the world worse off.

The review correctly points out that climate change is a real problem and that it is caused by human greenhouse-gas emissions. Little else is right, however, and the report seems hastily put together, with many sloppy errors. As an example, the cost of hurricanes in the US is said to be both 0.13 per cent of US gross domestic product and 10 times that figure.

The review is also one-sided, focusing almost exclusively on carbon-emission cuts as the solution to the problem of climate change. Stern sees increasing hurricane damage in the US as a powerful argument for carbon controls. However, hurricane damage is increasing predominantly because there are more people with more goods to be damaged, settling in more risky habitats.

Even if global warming does significantly increase the power of hurricanes, it is estimated that 95 per cent to 98 per cent of the increased damage will be due to demographics. The review acknowledges that simple initiatives such as bracing and securing roof trusses and walls can cheaply reduce damage by more than 80 per cent; yet its policy recommendations on expensive carbon reductions promise to cut the damage by 1 per cent to 2 per cent at best. That is a bad deal.

Stern is also selective, often seeming to cherry-pick statistics to fit an argument. This is demonstrated most clearly in the review's examination of the social damage costs of CO2, essentially the environmental cost of emitting each extra tonne of CO2.

The most well-recognised climate economist in the world is probably Yale University's William Nordhaus, whose "approach is perhaps closest in spirit to ours", according to the Stern review. Nordhaus finds that the social cost of CO2 is $2.50 per tonne. Stern, however, uses a figure of $85 per tonne. Picking a rate even higher than the official British estimates - which have been criticised for being over the top - speaks volumes.

Stern tells us that the cost of flooding in Britain will quadruple to 0.4 per cent from 0.1 per cent of GDP because of climate change. However, we are not told that these alarming figures only hold true if one assumes that Britain will take no additional measures, essentially doing absolutely nothing and allowing itself to get flooded, perhaps time and again.

In contrast, the British Government's assumptions take into account a modest increase in flood prevention, finding that the cost will decline sharply to 0.04 per cent of GDP, in spite of climate change. Why does Stern not share that information?

But nowhere is the imbalance clearer than in Stern's central argument about the costs and benefits of action on climate change. The review tells us that we should make significant cuts in carbon emissions to stabilise the concentration of atmospheric CO2 at 550 parts per million. Yet such a stark recommendation is not matched by an explicit explanation of what this would mean in terms of temperature.

The UN climate panel estimates that stabilising at 550ppm would mean an increase in temperature of about 2.3C by the year 2100. This might be several degrees below what would otherwise happen, but it might also be higher. Nordhaus estimates that the stabilisation policy would reduce the rise from 2.53C to 2.42C. One can understand the reluctance of the Stern review to advertise such a puny effect.

Most economists were surprised by Stern's large economic estimates of damage from global warming. Nordhaus's model, for example, anticipates 3 per cent will be wiped off global GDP if nothing is done over the coming century, taking into account the risk for catastrophes. The Stern review purports to show that the cost is "larger than many earlier studies suggested".

On the face of it, Stern accepts Nordhaus's figure: even including risks of catastrophe and non-market costs, he agrees that an increase of 4C will cost about 3 per cent of GDP. But he assumes that we will continue to pump out carbon far into the 22nd century, a rather unlikely scenario given the falling cost of alternative fuels, and especially if some of his predictions become clear to us towards the end of this century. Thus he estimates that the higher temperatures of 8C in the 2180s will be very damaging, costing 11 per cent to 14 per cent of GDP.

The Stern review then analyses what the cost would be if everyone in the present and the future paid equally. Suddenly the cost estimate is not 0 per cent now and 3 per cent in 2100, but 11 per cent of GDP right now and forever. If this seems like a trick, it is certainly underscored by the fact that the Stern review picks an extremely low discount rate, which makes the cost look much more ominous now.

But even 11 per cent is not the last word. Stern suggests that there is a risk that the cost of global warming will be higher than the top end of the UN climate panel's estimates, inventing, in effect, a worst-case scenario even worse than any others on the table. Therefore, the estimated damage to GDP jumps to 15 per cent from 11 per cent. Moreover, Stern admonishes that poor people count for less in the economic calculus, so he then inflates 15 per cent to 20 per cent.

This figure, 20 per cent, was the number that rocketed across the world, although it is simply a much-massaged reworking of the standard 3 per cent of GDP cost in 2100, a figure accepted among most economists to be a reasonable estimate.

Likewise, Stern readjusts the cost of dealing with climate change. The UN found that the cost of 550ppm stabilisation would be somewhere from 0.2 per cent to 3.2 per cent of GDP today; he reports that costs could lie between minus 4 per cent and 15 per cent of GDP. The minus 4 per cent is based on the suggestion that cutting carbon emissions could make us richer because revenue recycling could address inefficiencies in taxation; but the alleged inefficiencies, if correct, should be addressed no matter what the policies about climate change.

The reason Stern nevertheless finds a very low cost estimate is because he only considers models with so-called induced technological change. These models are known to reduce costs by about two percentage points because carbon cuts lead to an increase in research and development, which again makes further cuts cheaper. Thus, Stern concludes that the costs are on average 1 per cent of GDP, and in the summary claims that this is a maximum cost.

The Stern review's cornerstone argument for immediate and strong action is based on the suggestion that doing nothing about climate change costs 20 per cent of GDP now, and doing something only costs 1 per cent. However, this argument hinges on three very problematic assumptions.

First, it assumes that if we act, we will not still have to pay. But this is not so: Stern tells us that his solution is "already associated with significant risks". Second, it requires the cost of action to be as cheap as he tells us, and on this front his numbers are at best overly optimistic. Third, and most significantly, it requires the cost of doing nothing to be a realistic assumption, but the 20 per cent of GDP figure is inflated by an unrealistically pessimistic vision of the 22nd century, and by an extreme and unrealistically low discount rate. According to the background numbers in Stern's report, climate change will cost us 0 per cent now and 3 per cent of GDP in 2100, a much more informative number than the 20 per cent now and forever.

In other words, given reasonable inputs, most cost-benefit models show that dramatic and early carbon reductions cost more than the good they do. Stern's attempt to challenge that understanding is based on a chain of unlikely assumptions.

Moreover, there is a fourth key problem in Stern's argument that has received very little attention. It seems naive to believe that the world's 192 nations can flawlessly implement Stern's multi-trillion-dollar, century-long policy proposal. Will nobody try to avoid its obligations? Why would China and India participate? And even if China got on board, would it be able to implement the policies?

In 2002, China decided to cut sulphur dioxide emissions by 10 per cent; they are now 27 per cent higher despite SO2 being nationally a much bigger health and environmental problem than climate change.

We all want a better world. But we must not let ourselves be swept up in making a bad investment simply because we have been scared by sensationalist headlines.